Classification of Students' Mathematical Answers Using BART-Based Machine Learning Konjarla Sripriya, Likhitha Sri Kandukuri, Kammari Vidyasri, Roshni M Balakrishnan, Peeta Basa Pati Proceedings of 3rd IEEE International Conference on Knowledge Engineering and Communication Systems Ickecs 2025, 2025 Using machine learning (ML) for automated grading is a big step toward effective and reliable assessment in today’s educational environment. This work aims to use ML to evaluate student answers to various questions, such as simultaneous equations, word problems, and quadratic equations. Our method attempts to reliably classify answers as correct, partially correct, and incorrect by imitating expert-level grading. This research explores the use of BART embeddings using the application of Stacking Classifiers. XGBoost as a meta-classifier has exhibited the best accuracy of $97.89 \%$ mean accuracy with standard deviation of 0.02. The results demonstrate the potential of machine learning techniques to revolutionize the grading process, ensuring consistency and efficiency in educational assessments.
Unveiling Patterns in Student Problem-Solving: Semantic Clustering and Topic Modeling with T5 and BART Embeddings C Madhuja, Shreya Bhanot, Samreen Sidhu, Roshni M Balakrishnan, Peeta Basa Pati 6th International Conference on Control Communication and Computing Iccc 2025, 2025 For students, it is very important to know our strengths and weaknesses in mathematics so that we can improve overall performance. Traditional grading systems often provide only a summary of marks without offering insights into where we are losing them. This study proposes a transformative approach to assessments by leveraging transformer-based models, T5((Text-to-Text Transfer Transformer) and BART(Bidirectional and Auto-Regressive Transformer), to analyze and cluster student responses based on mathematical subdomains like trigonometry, geometry, and algebra. It will identify patterns in our problem-solving techniques by extracting semantic embeddings from our answers and applying clustering algorithms such as K-Means and Mean Shift. This insight will enable us to know the specific areas where mistakes are happening, thus pinpointing the subdomains that need improvement. A personalized learning experience will be achieved this way. Furthermore, an auto-recommendation system is foreseen to recommend targeted resources and strategies to enhance our skills in those areas. This work envisions assessments as something serving growth and self-awareness, fostering not just better academic performance but a deeper grasp of mathematical concepts through innovative machine learning applications.
DeepNeuroRest: Deep Learning-Assisted Stress Detection in Sleep with Smart Web App Insights Angelina George, Achal Baniya, Alphonsa Jose, Roshni M Balakrishnan, Peeta Basa Pati Proceedings of 3rd International Conference on Augmented Intelligence and Sustainable Systems Icaiss 2025, 2025 This study conducted a comprehensive analysis of stress level prediction using both machine learning and deep learning models. Phase 1 involved training nine machine learning models, including Decision Tree, Random Forest, Support Vector Machine, k-Nearest Neighbours, Naive Bayes, CatBoost, eXtreme Gradient Boosting, Gradient Boosting, and Extra Trees, alongside five deep learning models: Artificial Neural Networks, Long Short-Term Memory, Radial Basis Function, Gated Recurrent Unit, and Multi-Layer Perceptron. Initially, these models achieved perfect testing accuracy, precision, recall, and F1-Score on the original dataset. However, when incorporating fine-tuned Auxiliary Classifier Generative Adversarial Networks (AC-GAN) to enhance the dataset, the best machine learning performance was observed with CatBoost and Extra Trees, each achieving 98.77% accuracy and an F1-Score of 1 for Class 0, 0.99 for Classes 1 and 2, and 0.98 for Classes 3 and 4. The optimal deep learning model was the Multi-Layer Perceptron, which achieved a test accuracy of 97.852% and a minimal loss of 5.438%. Phase 2 involved developing two client applications to analyse stress levels: an email client and a web client. The best deep learning model was utilized for these applications. The email client, developed using Microsoft Azure, Flask Server, and Microsoft Automation, and the web client, built with Wix, Streamlit, and ngrok, both provided users with stress level analysis depicted through a speedometer graph. These implementations show how the models can be used practically for managing and monitoring stress levels in real-world situations.
Automatic Grading of Students’ Algebraic Responses Using MathBERT, GPT-2 & Auto-Tokenizer Embeddings Aryan Kothari, VRNS Nikhil, Namana Rohit, Roshni M Balakrishnan, Peeta Basa Pati Proceedings of 2025 International Conference on Emerging Technologies in Computing and Communication Etcc 2025, 2025 This study introduces a unique approach by applying dense vector representations specifically tailored to capture the nuances in algebraic answer submissions, addressing the challenges of linguistic and mathematical complexity in student responses. Unlike previous approaches focused on syntactic matching or simpler models, this research leverages Math-BERT embeddings-designed for mathematical language-and provides a thorough comparative analysis using GPT-2 and Auto-Tokenizer embeddings. Using expert-rated scores as benchmarks, alongside extensive hyperparameter tuning, this study delivers a data-driven solution for evaluating and grading algebraic answers. The results demonstrate MathBERT's effectiveness in capturing intricate mathematical reasoning beyond general-purpose language models, achieving high R-squared and low RMSE values. This methodology holds potential for enhancing grading accuracy and feedback quality in mathematical assessments.
Agri Assist: An AI Integrated Farmer Assistant Pennabadi Devendra Reddy, K Satya Sampath Reddy, P Jayanth, Bhanu Prakash Kakarla, Roshni M Balakrishnan Procedia Computer Science, 2025 In this paper, an AI Agriculture Assistant integrating a crop recommendation system and an agricultural query chatbot is developed to provide guidance on crop cultivation. Based on this, the crop recommendation system uses a stack ensemble model comprising Random Forest and Gradient Boosting, and an accuracy of 99.32% and F1-Score of 99.26%. Here, word embeddings are performed through FastText which allows for a quick response time (0.0244 seconds) and a cosine similarity of 0.88629 for the chatbot. With RSA encryption for securing user data, we ensure secure communication. It uses the critical elements of the soil composition, weather and crop performance in formulating tailored agricultural guidance. Back end communication is done through Flask, and the entire system was deployed in this manner with a web interface to drive usage and for real time data responses. This entire application helps the farmers in decision making regarding crops cultivation and management through accurate and secure assistance.
Analysis Of The Efficiency of MathBERT and T5 Embeddings in Classification of Students' Responses to Algebraic Questions based on their Correctness and Relevance Yashaswini Manyam, Srinidhi Sundaram, Roshni M Balakrishnan, Peeta Basa Pati 2024 5th IEEE Global Conference for Advancement in Technology Gcat 2024, 2024 Automated grading processes helps in saving time for teachers, which can be devoted to more instruction and other essential services. Behavioral analysis of ML models on students’ responses to algebraic questions solve problems that exist with traditional manual grading methods thereby easing the manual assessment process. The study involves the application of machine learning classifiers on the 384 dimensional feature extracted using MathBERT and 768 dimensional features extracted using T5 for each response to the algebraic questions in order to analyse their correctness and relevance. The algorithms are trained over the labeled dataset of a wide variety of algebraic answers and provided with the correct and incorrect answers to learn different patterns. The research aims at analyzing the efficiency of MathBERT and T5, which were trained on LaTeX representations of mathematical expressions and massive text, code and symbols data respectively, in generating embeddings that understand the underlying semantics of mathematical expressions.
Fine-Tuned T5 for Auto-Grading of Quadratic Equation Problems Roshni M Balakrishnan, Peeta Basa Pati, Rimjhim Padam Singh, Santhanalakshmi S, Priyanka Kumar Procedia Computer Science, 2024 Assessments constitute a fundamental and inevitable component of any educational journey. Manual effort required for the evaluation of these assessments is very high. Automation of the evaluation process and grading helps in making the review process more efficient, objective, and scalable, thereby reducing the workload of human reviewers. Automating the grading process for multiple-choice and short-answer assessments is relatively straightforward, but it poses significant challenges when applied to the evaluation of formal languages, particularly in the context of mathematical assessments. In this paper a model that automatically evaluates and grades the Quadratic Equation problems is presented. The study is conducted using a manually curated dataset comprising 1200 solutions to various quadratic equation problems. Embeddings of the quadratic solutions are generated using Google’s T5 Model. These embeddings are then used to train different traditional and ensembled machine learning models along with complex Deep learning models like LSTM and Bi LSTM. An in-depth analysis of the fine-tuned T5 model’s performance, evaluating its effectiveness in comparison with the pretrained T5 model in automatic grading of quadratic equation problems has been explored. Fine-tuning significantly contributes to the reduction of error by 70% and a noticeable increase in the R2 value to 97%.
AudioGuard: Deep Learning Based Telugu DeepFake Audio Detection Siwani Karna, Samudrala Sai Santhoshi Haneesha, Poluru Reddy Jahanve, Peeta Basa Pati, Roshni M Balakrishnan 2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
Automatic Grading of Students' Algebraic Responses Using MathBERT, GPT-2 & Auto-Tokenizer Embeddings A Kothari, V Nikhil, N Rohit, RM Balakrishnan, PB Pati 2025 International Conference on Emerging Technologies in Computing and … , 2025 2025
Unveiling Patterns in Student Problem-Solving: Semantic Clustering and Topic Modeling with T5 and BART Embeddings C Madhuja, S Bhanot, S Sidhu, RM Balakrishnan, PB Pati 2025 6th International Conference on Control, Communication and Computing … , 2025 2025
DeepNeuroRest: Deep Learning-Assisted Stress Detection in Sleep with Smart Web App Insights A George, A Baniya, A Jose, RM Balakrishnan, PB Pati 2025 Third International Conference on Augmented Intelligence and … , 2025 2025
Classification of Students’ Mathematical Answers Using BART-Based Machine Learning K Sripriya, LS Kandukuri, K Vidyasri, RM Balakrishnan, PB Pati 2025 International Conference on Knowledge Engineering and Communication … , 2025 2025
Agri Assist: An AI Integrated Farmer Assistant PD Reddy, KSS Reddy, P Jayanth, BP Kakarla, RM Balakrishnan Procedia Computer Science 258, 3510-3522 , 2025 2025 Citations: 10
Text Line Segmentation of Palm Leaf Documents Using Segformer U Karamalapudi, NKS Keerthan, NK Musunuru, RM Balakrishnan, ... International Conference on Communication and Intelligent Systems, 389-404 , 2024 2024
Prompt Palate: A Prompt-Based Restaurant Recommendation System with Scheduling and Chatbot Integration NA Vidula, TH Babu, VG Kiran, RM Balakrishnan, S Bhaskaran 2024 8th International Conference on Computational System and Information … , 2024 2024
Analysis of the efficiency of MathBERT and T5 embeddings in classification of students’ responses to algebraic questions based on their correctness and relevance Y Manyam, S Sundaram, RM Balakrishnan, PB Pati 2024 5th IEEE Global Conference for Advancement in Technology (GCAT), 1-7 , 2024 2024 Citations: 1
Nova: Revolutionizing agile project management with scrum and kanban integration for enhanced team collaboration and productivity P Manaswini, S Shaik, S Ashwani, RM Balakrishnan, SM Rajagopal 2024 5th IEEE Global Conference for Advancement in Technology (GCAT), 1-7 , 2024 2024 Citations: 6
Harnessing the Power of web3: A Blockchain Approach to Crowdfunding RV Savant, SN Sunder, M Spoorthi, SM Rajagopal, RM Balakrishnan 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 2
Automatic Assessment of Quadratic Equation Solutions Using MathBERT and RoBERTa Embeddings SS Rao, S Mishra, S Akhilesh, RM Balakrishnan, PB Pati 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 3
Audioguard: Deep learning based telugu deepfake audio detection S Karna, SSS Haneesha, PR Jahanve, PB Pati, RM Balakrishnan 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 8
A Machine Learning Based Auto-Grading Model for Question-Based Algebra Problems using RoBERTa NA Vidula, TA Reddy, S Manogna, RM Balakrishnan, PB Pati 2024 4th International Conference on Intelligent Technologies (CONIT), 1-6 , 2024 2024 Citations: 5
A machine learning based classification of students' algebraic responses using mathbert embeddings AS Injeti, GN Rupsica, GP Reddy, RM Balakrishnan, PB Pati 2024 5th International Conference for Emerging Technology (INCET), 1-6 , 2024 2024 Citations: 4
Cluster Based Classification of Question Independent C Codes A Vijjapu, AH Meti, A Rao, RM Balakrishnan, PB Pati International Conference on Computer, Communication, and Signal Processing … , 2024 2024 Citations: 1
Fine-tuned T5 for auto-grading of quadratic equation problems RM Balakrishnan, PB Pati, RP Singh, P Kumar Procedia Computer Science 235, 2178-2186 , 2024 2024 Citations: 12
Oral Squamous Cell Carcinoma Diagnosis Using Spotted Hyena Optimizer Combined with Transfer Learning Approaches RM Balakrishnan, N Bharanidharan, SRS Chakravarthy, VV Kumar, ... International Conference on Soft Computing and Pattern Recognition, 45-53 , 2023 2023
Dimensionality reduction on fish dataset with different machine learning algorithms comparison C Sugunadevi, RM Balakrishnan 2023 International Conference on Ambient Intelligence, Knowledge Informatics … , 2023 2023 Citations: 2
Empowering Speech-Impaired Individuals: EEG-Driven Cognitive Expression Translated into Speech J Vrindavanam, RM Balakrishnan, R Nanjappan, G Kamath International Journal of Computer Applications 185 (28), 43-46 , 2023 2023
Auto-grading C programming assignments with CodeBERT and Random Forest Regressor RV Muddaluru, SR Thoguluva, S Prabha, PB Pati, RM Balakrishnan 2023 14th International Conference on Computing Communication and Networking … , 2023 2023 Citations: 24
MOST CITED SCHOLAR PUBLICATIONS
Auto-grading C programming assignments with CodeBERT and Random Forest Regressor RV Muddaluru, SR Thoguluva, S Prabha, PB Pati, RM Balakrishnan 2023 14th International Conference on Computing Communication and Networking … , 2023 2023 Citations: 24
Fine-tuned T5 for auto-grading of quadratic equation problems RM Balakrishnan, PB Pati, RP Singh, P Kumar Procedia Computer Science 235, 2178-2186 , 2024 2024 Citations: 12
Agri Assist: An AI Integrated Farmer Assistant PD Reddy, KSS Reddy, P Jayanth, BP Kakarla, RM Balakrishnan Procedia Computer Science 258, 3510-3522 , 2025 2025 Citations: 10
Audioguard: Deep learning based telugu deepfake audio detection S Karna, SSS Haneesha, PR Jahanve, PB Pati, RM Balakrishnan 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 8
Nova: Revolutionizing agile project management with scrum and kanban integration for enhanced team collaboration and productivity P Manaswini, S Shaik, S Ashwani, RM Balakrishnan, SM Rajagopal 2024 5th IEEE Global Conference for Advancement in Technology (GCAT), 1-7 , 2024 2024 Citations: 6
A Machine Learning Based Auto-Grading Model for Question-Based Algebra Problems using RoBERTa NA Vidula, TA Reddy, S Manogna, RM Balakrishnan, PB Pati 2024 4th International Conference on Intelligent Technologies (CONIT), 1-6 , 2024 2024 Citations: 5
Theoretical analysis of expected population variance evolution for a differential evolution variant S Thangavelu, G Jeyakumar, RM Balakrishnan, CS Velayutham Computational Intelligence in Data Mining-Volume 2: Proceedings of the … , 2014 2014 Citations: 5
A machine learning based classification of students' algebraic responses using mathbert embeddings AS Injeti, GN Rupsica, GP Reddy, RM Balakrishnan, PB Pati 2024 5th International Conference for Emerging Technology (INCET), 1-6 , 2024 2024 Citations: 4
Automatic Assessment of Quadratic Equation Solutions Using MathBERT and RoBERTa Embeddings SS Rao, S Mishra, S Akhilesh, RM Balakrishnan, PB Pati 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 3
Harnessing the Power of web3: A Blockchain Approach to Crowdfunding RV Savant, SN Sunder, M Spoorthi, SM Rajagopal, RM Balakrishnan 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 2
Dimensionality reduction on fish dataset with different machine learning algorithms comparison C Sugunadevi, RM Balakrishnan 2023 International Conference on Ambient Intelligence, Knowledge Informatics … , 2023 2023 Citations: 2
Analysis of the efficiency of MathBERT and T5 embeddings in classification of students’ responses to algebraic questions based on their correctness and relevance Y Manyam, S Sundaram, RM Balakrishnan, PB Pati 2024 5th IEEE Global Conference for Advancement in Technology (GCAT), 1-7 , 2024 2024 Citations: 1
Cluster Based Classification of Question Independent C Codes A Vijjapu, AH Meti, A Rao, RM Balakrishnan, PB Pati International Conference on Computer, Communication, and Signal Processing … , 2024 2024 Citations: 1
Automatic Grading of Students' Algebraic Responses Using MathBERT, GPT-2 & Auto-Tokenizer Embeddings A Kothari, V Nikhil, N Rohit, RM Balakrishnan, PB Pati 2025 International Conference on Emerging Technologies in Computing and … , 2025 2025
Unveiling Patterns in Student Problem-Solving: Semantic Clustering and Topic Modeling with T5 and BART Embeddings C Madhuja, S Bhanot, S Sidhu, RM Balakrishnan, PB Pati 2025 6th International Conference on Control, Communication and Computing … , 2025 2025
DeepNeuroRest: Deep Learning-Assisted Stress Detection in Sleep with Smart Web App Insights A George, A Baniya, A Jose, RM Balakrishnan, PB Pati 2025 Third International Conference on Augmented Intelligence and … , 2025 2025
Classification of Students’ Mathematical Answers Using BART-Based Machine Learning K Sripriya, LS Kandukuri, K Vidyasri, RM Balakrishnan, PB Pati 2025 International Conference on Knowledge Engineering and Communication … , 2025 2025
Text Line Segmentation of Palm Leaf Documents Using Segformer U Karamalapudi, NKS Keerthan, NK Musunuru, RM Balakrishnan, ... International Conference on Communication and Intelligent Systems, 389-404 , 2024 2024
Prompt Palate: A Prompt-Based Restaurant Recommendation System with Scheduling and Chatbot Integration NA Vidula, TH Babu, VG Kiran, RM Balakrishnan, S Bhaskaran 2024 8th International Conference on Computational System and Information … , 2024 2024
Oral Squamous Cell Carcinoma Diagnosis Using Spotted Hyena Optimizer Combined with Transfer Learning Approaches RM Balakrishnan, N Bharanidharan, SRS Chakravarthy, VV Kumar, ... International Conference on Soft Computing and Pattern Recognition, 45-53 , 2023 2023
